| Abstract |
VeLASSCo will provide new visual analysis methods for large-scale simulations serving the petabyte era and preparing the exabyte era by adopting Big Data tools/architectures for the engineering and scientific community, leveraging new ways of in-situ processing for data analytics and hardware accelerated interactive visualization.
Regardless of its origin, in the near future the challenge will not be on how to generate data, but rather in how to manage big and highly distributed data to make it easily handled by users in their personal devices. In the case of simulation data, the extraction of the specific useful information for analysis requires the treatment of large amounts of the data generated by the simulation, such as data for each time step, entire volume results when only some cuts are needed, etc.
The global goal of VELaSSCo is to provide Big Data tools, for the engineering and scientific community in order to better manipulate simulation with billions of distributed records.
In large-scale simulations the domain is partitioned across several thousands of nodes, and the data (mesh and results) is stored on those nodes in a distributed manner. The VELaSSCo platform to be developed in this project will access this distributed information, will process it and send the results back to the users for local visualization by their specific visualization clients and tools (GiD from CIMNE, iFX from Fraunhofer, or other).
Based on the layered structure commonly used in the Big Data field, the architecture of VELaSSCo platform is based on three layers:
- Service layer: it is the communication point between the platform and the user's visualization client, responsible of classifying user queries and send the results back. It will ensure access to the distributed data and provide a unified view of it.
- Speed layer: it will ensure nearly-instant access and visualization of large-scale data, so that the user can interact with the simulation results (sub-second response times would be achieved by using simplified version of the data).
- Batch layer: it will perform the heavy data processing tasks, such as complex queries, creation of simplified versions of the data and interpolation of DEM data to continuum domain. This layer will ensure resolution of complex queries and provide processed data to the speed layer.
VELaSSCo will take into account the different types of simulation schemes (FEM, DEM, LB, etc.) and fields of application (Fluid and solid mechanics, electromagnetism, etc.) that could benefit from the use of its platform.
The VELaSSCo consortium includes experts with relevant background in Big Data handling, advanced visualisation, engineering and scientific simulations. The team will be reinforced with the experience and feedback of a User Panel including research centres, SMEs and companies form key European industrial sectors such as aerospace, household products, chemical, pharmaceutical and civil engineering. |
